OUP user menu

Novel strategies for the prevention and treatment of Candida infections: the potential of immunotherapy

Frank L. Van De Veerdonk, Mihai G. Netea, Leo A. Joosten, Jos W.M. Van Der Meer, Bart Jan Kullberg
DOI: http://dx.doi.org/10.1111/j.1574-6976.2010.00232.x 1063-1075 First published online: 1 November 2010

Abstract

Infections caused by Candida spp. continue to be a substantial cause of disease burden, especially in immunocompromised patients. New approaches are needed to improve the outcome of patients suffering from Candida infections, because it seems unlikely that the established standard treatment will drastically lower the morbidity of mucocutaneous Candida infections and the high mortality associated with invasive candidiasis. New insights into the mechanisms of the anti-Candida host response have contributed to the design of novel immunotherapeutic approaches that have been proposed as adjuvant therapy in Candida infections. This review presents an overview of novel strategies in the prevention and treatment of Candida infections, with a special focus on adjuvant immunotherapy.

Keywords
  • Candida
  • infections
  • treatment
  • immunotherapy

Introduction

Candida infections can be divided into local infections and invasive candidiasis. The various forms of mucosal Candida infections induce a significant burden of morbidity: vulvovaginal candidiasis (VVC), oropharyngeal candidiasis (OPC) and chronic mucocutaneous candidiasis (CMC). VVC, the most common form of mucosal candidiasis, is widespread and may affect up to 75% of women of child-bearing age (Cassone et al., 2007). It is characterized by pruritus, irritation and dyspareunia, often accompanied by increased vaginal discharge. VVC can be divided into uncomplicated or complicated (Sobel et al., 1998). Complicated VVC is defined as severe or recurrent disease, infection due to Candida spp. other than Candida albicans or VVC in an abnormal host (Sobel et al., 1998). Recurrent forms of VVC (RVVC) are defined as four or more episodes of symptomatic VVC within 1 year (Sobel et al., 2004). Notably, in patients with RVVC, the associated costs of medical visits are high and their quality of life is significantly reduced (Cassone et al., 2007).

OPC is a relatively common local infection occurring in high-risk groups of patients such as dental wearers, diabetic patients, individuals treated with broad-spectrum antibiotics, infants and patients infected with HIV. It has been reported that 84% of HIV-infected individuals are asymptomatically colonized with Candida spp. in the oral cavity, with 55% developing at least one episode of OPC with clinical signs (Sangeorzan et al., 1994). If left untreated, these lesions contribute considerably to the morbidity associated with HIV infection (Powderly et al., 1999). Importantly, OPC can be complicated by esophageal candidiasis. Esophageal candidiasis is accompanied by more serious complaints and predisposes to the development of systemic candidiasis (Richet et al., 1991; Samonis & Bafaloukos, 1992).

These mucosal manifestations of candidiasis may also be associated with primary immunodeficiencies. CMC represents a heterogeneous group of primary immunodeficiencies that are characterized by an inability to clear fungal infections. Consequently, persisting and recurrent infections of the skin and mucous membranes with C. albicans ensue (Lilic, 2002). Some patients have autosomal recessive polyglandular autoimmune syndrome type I, also known as the autoimmune polyendocrinopathy candidiasis ectodermal dystrophy syndrome. This syndrome is caused by mutations in the autoimmune regulator gene and is characterized by CMC and endocrine disorders, such as hypoparathyroidism and Addison's disease (Ahonen et al., 1990). CMC also occurs without associated disorders and the genetic defect in these patients is not known (Kirkpatrick et al., 2001). Another primary immunodeficiency accompanied by mucocutaneous candidiasis is the hyperimmunoglobulin E syndrome (HIES, Job's syndrome). This disease is further characterized by recurrent staphylococcal skin abscesses, pulmonary infections, skeletal and dental abnormalities, and elevated serum immunoglobulin E concentrations (Davis et al., 1966; Grimbacher et al., 1999). Recently, it has been demonstrated that in a large number of patients with HIES, a dominant-negative mutation in STAT3 is the underlying cause of disease (Minegishi et al., 2007).

Disseminated candidiasis is a deep-seated organ infection with Candida spp. and/or candidemia. The clinical spectrum varies from minimal fever to severe sepsis with multiorgan failure (Guery et al., 2009). The microorganism gains access to the intravascular compartment either from the gastrointestinal tract or, less often, from the skin through the site of an indwelling intravascular catheter. Noteworthy, it has become apparent that health care workers' hands are commonly colonized with Candida spp. (Brunetti et al., 2008), and several studies have demonstrated that this is the main cause of the reported Candida parapsilosis outbreaks in pediatric intensive care unit (ICU) departments (Huang et al., 1999; Lupetti et al., 2002; Hernandez-Castro et al., 2009). Candida spp. has been reported as one of the most common pathogens that cause hospital-acquired bloodstream infections in patients undergoing surgical or chemotherapeutic interventions and/or with underlying immunological deficiencies (Wisplinghoff et al., 2004; Pfaller & Diekema, 2007). Importantly, nosocomial candidemia is associated with an unacceptably high mortality ranging between 30% and 50% (Gudlaugsson et al., 2003; Wisplinghoff et al., 2004; Zaoutis et al., 2005).

In this review, we will discuss novel approaches such as vaccination, antibodies, cytokine therapy and adoptive transfer of primed immune cells that have the potential to improve the clinical outcome of patients with Candida infections.

Diagnosis and treatment of Candida infections

Patients with invasive candidiasis should receive effective treatment as soon as possible, because delays in starting antifungal therapy in candidemia are correlated with increased mortality (Garey et al., 2006; Guery et al., 2009; Pappas et al., 2009). Therefore, it is of utmost importance to identify candidemia in critically ill patients as soon as possible. Several clinical scores have been developed to identify patients who are at a high risk of developing candidemia (Leon et al., 2006; Ostrosky-Zeichner et al., 2007). Microbiological identification of the yeast in the blood or the organs remains the gold standard for the diagnosis of disseminated candidiasis. Unfortunately, the microbiological methods have a low sensitivity, and additional methods have been developed to improve early fungal detection. Noninvasive techniques, such as the measurement of mannan and anti-mannan antibodies and the detection of the fungal cell wall component (1–3)-βd-glucan, have the potential to be useful for the early diagnosis of invasive candidiasis (Prella et al., 2005; Kedzierska et al., 2007; Ellis et al., 2009). Another promising technique is the use of a PCR, which has been reported to have a high sensitivity and specificity and has good potential to provide species identification before blood culture positivity (Lau et al., 2009). Still, none of these techniques can definitively prove a Candida infection (Ellepola & Morrison, 2005; Zaragoza et al., 2009). Because it remains difficult to establish a diagnosis of invasive candidiasis early in the course of infection, prophylactic therapy is often advocated. However, prophylactic treatment in high-risk groups, such as stem cell transplant recipients and high-risk populations in the ICU, is a controversial issue as outlined in the IDSA guidelines for the treatment of invasive candidiasis (Pappas et al., 2009).

The diagnosis of VVC is usually made on the basis of clinical signs and symptoms (Cassone et al., 2007), but because these are not specific, it is wise to confirm it with a smear before starting antifungal therapy. Clinical signs and symptoms of OPC in HIV patients are usually suggestive of the diagnosis (de Repentigny et al., 2004). However, in view of the differential diagnosis, frequent recurrence or persistence, microbiological culture to confirm the diagnosis of OPC, speciation and susceptibility testing is required.

The current standard treatment of Candida infections consists of antifungal agents such as azoles, echinocandins and amphotericin B compounds (Box 1). Despite these available antifungal agents, the frequency of treatment failure is considerable, underscoring the necessity for new treatment strategies. In vitro experiments suggest that combinations of these antifungal agents might improve antifungal efficacy (Baltch et al., 2008; Tobudic et al., 2010). However, no clinical trial to date has demonstrated that combining antifungal agents results in additional efficacy for the treatment of invasive candidiasis. The lack of progress in terms of mortality due to disseminated Candida infections has led to the opinion among the experts that only a combination of standard antifungal treatment and adjunctive immunotherapy will be able to decrease the mortality in Candida sepsis. A better understanding of the pathogenesis of host defense against Candida infections has made it possible to explore immunomodulatory interventions that might contribute to the standard clinical practice.

View this table:
Box 1.

Standard treatment of Candida infections

Invasive candidiasis
• For neutropenic patients, patients with moderate to severe illness or hemodynamic unstable patients, those with previous azole exposure and/or patients who are at risk of infection by C. glabrata or C. krusei, echinocandins are recommended as the primary treatment (Pappas et al., 2009). Other patients may receive fluconazole treatment.
• Once the patient is clinically stable, the Candida species is identified and its susceptibility is known, echinocandins may be switched to fluconazole or voriconazole and conversion to oral therapy should be considered.
• Treatment should be continued for 2 weeks after the last positive blood culture. If metastatic foci have occurred, the treatment duration should be prolonged.
Vulvovaginal candidiasis
• Successful treatment may be achieved with an oral or a topical agent (Sobel et al., 1998; Sobel et al., 2007), as these are equally effective (Watson et al., 2002). However, most often, treatment does not prevent recurrences (Cassone et al., 2007).
• Repeated treatment might select and induce drug resistance and a shift toward more resistant Candida species (Richter et al., 2005; Eckert et al., 2006).
Oropharyngeal candidiasis
• For mild disease, topical therapy is recommended, and for moderate to severe disease, oral fluconazole is recommended (Pappas et al., 2009).
• In HIV-infected patients, chronic suppressive therapy is usually unnecessary. The use of highly active anti-retroviral therapy (HAART) is recommended to reduce recurrent infections (Pappas et al., 2009).
Mucocutaneous candidiasis
• Fluconazole should be used as the first-line treatment in patients with CMC or HIES and most patients require chronic suppressive antifungal therapy (Pappas et al., 2009). This is due to the extensive skin or nail involvement and the many relapses often seen in these patients.

Pathogenesis of host defense against Candida infection

Host defense against Candida infection depends on intact mucosal and skin barriers, and adequate recognition of the fungus that subsequently triggers protective innate and adaptive antifungal defense mechanisms (Netea et al., 2008). The first line of defense is the skin and mucosa, which not only offers a mechanical barrier but also provides microbial antagonism with its associated normal flora. Once this first line of defense fails, protective innate and adaptive immune mechanisms will be activated that critically depend on appropriate pathogen recognition. Recognition of Candida is mediated by pattern recognition receptors (PRRs), which bind pathogen-associated molecular patterns (PAMPs). The most-studied and well-known PAMPs of Candida are components of the fungal cell wall. The inner layer of the cell wall is composed of β-(1,3)-glucan covalently linked to β-(1,6)-glucan (Iorio et al., 2008) and chitin, and an outer layer that consists of proteins that are heavily glycosylated by N-linked (Cutler et al., 2001) and O-linked mannosylation (Ernst & Prill, 2001). All these components have been reported as ligands for one or more PRRs, among which two classes collaborate in Candida recognition: toll-like receptors (TLRs), such as TLR2 and TLR4, and the C-type lectin receptors, such as dectin-1, dectin-2, Mincle and mannose receptor (MR) (Netea et al., 2002, 2006; Gow et al., 2007; Murciano et al., 2007; Wells et al., 2008; Robinson et al., 2009; van de Veerdonk et al., 2009b). Some of these receptors exert a proinflammatory effect (e.g. TLR4, dectin-1, MR), while others exert a more anti-inflammatory effect (e.g. TLR2) (Netea et al., 2008). The type of response initiated by Candida depends on the complex interaction between the PRRs expressed by the different cell types present at the site of infection (van de Veerdonk et al., 2008). Polymorphonuclear cells (PMNs), monocytes and macrophages are important for the main innate effector response: phagocytosis of Candida and the induction of reactive oxygen species (ROS) that can both damage and subsequently eliminate the fungus.

Proinflammatory cytokines such as tumor necrosis factor α (TNFα) and interleukin (IL)-1β are crucial for the proper activation of PMNs. TNFα is essential for anti-Candida host defense through the recruitment of neutrophils and phagocytosis, and deficiency results in higher mortality during experimental disseminated candidiasis (Netea et al., 2004). In addition, IFNγ produced by CD4 T lymphocytes is also important for the stimulation of the antifungal activity of PMNs. IFNγ induces NO production by macrophages and Candida-specific immunoglobulin production (Kaposzta et al., 1998). The central role of endogenous IFNγ in the resistance against systemic candidiasis has been underscored by the observation that knockout mice deficient in IFNγ are highly susceptible to C. albicans infection (Balish et al., 1998; Lavigne et al., 1998). Mice deficient in the cytokine IL-18, which plays a crucial role in the induction of IFNγ, are also more susceptible to disseminated candidiasis (Netea et al., 2003). Thus, the pathogenesis of invasive candidiasis seems to be linked to defects in phagocytosis and killing of Candida and defects, leading to IFNγ production (Fig. 1).

Figure 1

Pathogenesis of host defense against Candida. The endothelium and epithelium form the first line of defense. When Candida invades the tissues, innate immune cells will phagocytose and kill the fungus. This innate immune response can be followed by an adaptive response, which is initiated by cells that present antigen of the fungus. Naive T cells will be activated, and the cytokine profile that is present in the microenvironment will polarize the T helper response toward a predominant Th1, Th2 or Th17 response. IL=interleukin; IFN=interferon.

Epithelial cells are another source for the production of proinflammatory cytokines in the mucosa, and they have been advocated to play a central role in the protection against fungal pathogens (Mostefaoui et al., 2004; Dongari-Bagtzoglou & Fidel, 2005). Epithelial cells can produce IL-8 and granulocyte-macrophage colony-stimulating factor (GM-CSF) in response to Candida spp. (Dongari-Bagtzoglou & Kashleva, 2003; Pivarcsi et al., 2003; Li & Dongari-Bagtzoglou, 2009). Furthermore, oral epithelial cells can upregulate the antifungal activity of neutrophils in vitro, and this effect was partially dependent on IL-1α (Dongari-Bagtzoglou et al., 2005). In addition, neutrophils can upregulate TLR4 expression on C. albicans-infected human oral epithelium, and this was directly associated with protection against fungal invasion of the epithelium (Weindl et al., 2007).

The adaptive immune responses that are crucial for antifungal protection are elicited by CD4+ T helper (Th)1 (IFNγ-producing) cells and Th17 (IL-17/IL-22-producing) cells. The Candida-specific Th1 response is induced by antigen presentation in the presence of the cytokine IL-12 (Trinchieri et al., 1995), while Th17 responses are induced and maintained in the presence of IL-1 and IL-23 (Miossec et al., 2009). In patients with HIV who have low CD4 counts, the incidence of OPC is high (Ohmit et al., 2003). This underscores the importance of the T helper cell in mucosal anti-Candida host defense. Notably, in the absence of CD4+ T cells, CD8+ T cells also appear to play an important role in anti-Candida host defense (Myers et al., 2003; Marquis et al., 2006).

The recently discovered T helper subset Th17 cells have provided important novel insights into the pathogenesis of mucosal Candida infections. Th17 cells have been demonstrated to play a crucial role in host defense in experimental oropharyngeal Candida infection and disseminated candidiasis (Huang et al., 2004; Conti et al., 2009). It has been shown that patients with HIES have a defect in the Candida-induced Th17 response (Milner et al., 2008; van de Veerdonk et al., 2009a). Furthermore, in patients with CMC, there has been a link to a defect in their Th17 response against Candida (Eyerich et al., 2008). A recent study shows that IL-17A-deficient mice were equally susceptible to disseminated candidiasis, but in the same study, the Th17 response induced by vaccination was associated with protection against disseminated candidiasis (Lin et al., 2009). However, Th17 responses have also been suggested to be detrimental for the host during fungal infection. In an experimental fungal infection model, both inflammation and infection were exacerbated by the Th17 response against C. albicans and Aspergillus fumigatus (Zelante et al., 2007). This is an important controversy that needs further investigation, because these observations will provide the rationale for choosing the correct adjuvant in fungal vaccine strategies.

There have been quite a few speculations over the years as to whether patients with VVC have an underlying host defense defect (Cassone et al., 2007). It has been reported that the increase in vaginal mannose-binding lectin (MBL) levels in patients with VVC may be an effective immune response against C. albicans infection, and that women with recurrent VVC have lower vaginal levels of MBL compared with controls (Liu et al., 2006). In addition, MBL polymorphisms have been associated with recurrent VVC (Babula et al., 2003; Giraldo et al., 2007). Another polymorphism in the IL-4 (T-589) gene was found to correlate with a high prevalence of RVVC (Babula et al., 2005). One study has reported a role for neutrophils in the inflammatory response during VVC. Neutropenia, in an experimental model of vaginal Candida infection, had no effect on the fungal load during infection, but was significantly associated with decreased vaginal inflammation (Black et al., 1998). Another study reported that symptoms of VVC appear to be due to an aggressive innate response by PMN (Fidel et al., 2004).

Recently, new insights into the pathogenesis of RVVC have emerged. A family in which three women were affected by either recurrent VVC or onychomycosis were found to be dectin-1 deficient (Ferwerda et al., 2009). It was demonstrated that dectin-1 deficiency resulted in lower IL-17 production in response to C. albicans and this most probably accounts for the clinical picture seen in these patients. These observations were further strengthened by another report that describes a family that was deficient in CARD9, a downstream molecule in the signaling cascade of dectin-1 (Glocker et al., 2009). Patients with CARD9 deficiency had an impaired Candida-specific Th17 response, and female patients had a long history or an early onset of vaginal candidiasis. These data suggest that Th17 responses are important for vaginal mucosal host defense.

Novel immunotherapeutic strategies

Insights into the anti-Candida host defense mechanisms have contributed to the development of immune interventions that have the potential to lower the morbidity and mortality associated with Candida infections. Immunomodulatory strategies under investigation range from vaccination to therapeutic antibodies, recombinant cytokines and adoptive transfer of primed immune cells (Table 1).

View this table:
Table 1

Advantages and limitations of immunotherapeutic approaches

Immune strategyAdvantagesLimitationsAgentTherapeutic potential
VaccinationBroad-spectrum activity Low risk for the development of resistance Long-term effectsDependent on the competence of host immune status; most patients with disseminated candidiasis are immunocompromised Might also induce disease-enhancing Ab Relatively slow effectsRibosomal cell fraction (D.561)VVC (Levy et al., 1989)
Diphtheria toxoid CRM 197 conjugated with LamVVC; disseminated candidiasis (Torosantucci et al., 2005)
Mannan protein conjugatesVVC; disseminated candidiasis (Han et al., 1999)
Adhesins Als1p and Als3pOropharyngeal candidiasis (Spellberg et al., 2006); disseminated candidiasis (Ibrahim et al., 2006)
Als3p conjugated with alumDisseminated candidiasis (Lin et al., 2009)
Live attenuated Candida strain CA2Disseminated candidiasis (Bistoni et al., 1986)
Antibodies (Ab)Can be directly fungicidal independent of host immune status Very specific (e.g. strain specific) Rapid effects Could be used as monotherapyMay induce the development of anti-antibodies Potentially toxicRecombinant Ab against heat shock protein 90 (Efungumab)Invasive candidiasis (Pachl et al., 2006; Hodgetts et al., 2008)
Anti-mannan AbVVC (Han et al., 1999); Disseminated candidiasis (Han & Cutler, 1995; Han et al., 1999; Cutler et al., 2001)
Anti-β-glucan AbVVC (Pietrella et al., 2010); disseminated candidiasis (Torosantucci et al., 2005)
Idiotypic antibodies (YKT neutralizing Ab KT4)VVC (Polonelli et al., 1994); disseminated candidiasis (Polonelli et al., 1993)
CytokinesExperience with efficacy and safety in patients Already on the market Rapid effectsMainly effective as adjunctive therapy Might influence the pharmacokinetics and efficacy of the combined antifungal drug Inflammation as a potential consequenceGM-CSFRefractory mucosal candidiasis (Vazquez et al., 1998, 2000; Shahar et al., 1999); disseminated candidiasis (Lechner et al., 1994; Rokusz et al., 2001; Dignani et al., 2005)
G-CSFDisseminated candidiasis (Graybill et al., 1995; Kullberg et al., 1998, 2004)
IFNγOropharyngeal (in HIV) candidiasis (Bodasing et al., 2002); disseminated candidiasis (Kullberg et al., 1993; Dignani et al., 2005)
Adoptive transfer of primed immune cellsOpportunity to develop immune cells before immunocompromised statusRisk for developing graft-vs.-host disease after transplantationAdoptive transfer of anti-Candida T cells or pulsed dendritic cellsDisseminated candidiasis (Bacci et al., 2002; Tramsen et al., 2007)

Vaccination

Among those with invasive Candida infection, immunocompromised patients have the highest morbidity and mortality. Because vaccination depends on an appropriate host defense mechanism to provide protection, active and passive immunization in immunocompromised patients remains a challenge. Currently, no Candida vaccines are clinically available. Nevertheless, several active and passive fungal vaccine approaches appear to be promising and could prove to be an effective and safe strategy. A preliminary phase II trial with the oral vaccine D.561 performed already two decades ago in 22 patients with frequent recurrences of VVC showed promising results, but the true efficacy of the vaccine still has to be confirmed in a larger placebo-controlled trial (Levy et al., 1989). Several vaccine strategies have been tested in animal models with success. Diphtheria toxoid CRM197 conjugated with the algal antigen laminarin (Lam) was protective against both mucosal and systemic candidiasis in mice (Torosantucci et al., 2005). Mannan protein conjugates induced protective antibody responses against experimental disseminated candidiasis and Candida vaginal infection (Han et al., 1999). Vaccines based on the adhesins Als1p and Als3p were shown to induce a marked improvement in the fungal burden and survival of immunocompetent and immunocompromised mice with invasive and mucosal Candida infection (Ibrahim et al., 2005, 2006; Spellberg et al., 2005, 2006). Furthermore, vaccination with live-attenuated Candida or the low virulent CA2 strain has been reported to provide protection against hematogenous Candida reinfection in animal models (Bistoni et al., 1986). Although these different vaccination strategies yield promising results in animal models with experimental infections, their clinical safety and efficacy remains to be assessed in humans with Candida infections.

Important new insights have been gained in recent years regarding adjuvanticity. There are potent adjuvants that trigger the PRRs of dendritic cells (DCs), which results in DC maturation. Different adjuvants induce different cytokine profiles, and dependent on this profile, the immune response is shifted toward a Th1, Th2 or Th17 response. This enables to shape the kind of adaptive immune response elicited by the vaccine (Fig. 2). Several adjuvants available act through PRRs or their pathways, such as the TLR9 agonist CpG DNA (McCluskie & Krieg, 2006), the TLR4 agonist monophosphoryl lipid A (lipid A) (Mata-Haro et al., 2007) and alum, which exerts its effects through cryopyrin (NALP3) (Eisenbarth et al., 2008). To improve the efficacy of current vaccines, these insights should be exploited to specifically induce optimal defense against a pathogen. In candidiasis, this would mean that the adjuvant ideally facilitates a strong Th1 response in the case of disseminated candidiasis and a predominant Th17 response during mucosal candidiasis. The dectin-1 ligand β-glucan that has the potential to induce both a Th1 and a Th17 response would be a good candidate for a Candida vaccine adjuvant (Leibundgut-Landmann et al., 2008). Mannans from yeasts that induce a Th17 response through the MR and/or dectin-2 are of potential interest for vaccination against mucosal Candida infections (Robinson et al., 2009; van de Veerdonk et al., 2009a, b). A recent study has demonstrated that mice immunized with Als3p vaccine plus alum as an adjuvant were protected against disseminated candidiasis (Lin et al., 2009). Vaccination primed Th1 and Th17 lymphocytes, which resulted in neutrophil recruitment and activation at the site of infection and more effective clearance of C. albicans from the tissues.

Figure 2

Adjuvants and induction of specific Th1 and or Th17 responses. Overview of possible ligands (*) and their pathways that will result in a polarized T helper (Th) cell response. TLR=Toll-like receptor; IL=interleukin; IFN=interferon; ROR-c=RAR-related orphan receptor C; T-bet=T-box expressed in T cells; Syk=Spleen tyrosine kinase; Myd88=Myeloid differentiation primary response gene (88), Fcγ R=Fcγ receptor.

Antibodies

Anti-Candida antibodies induced artificially and administered to patients can be protective and might have the potential to be used as immunotherapy. This is evident from the literature that has investigated the use of antibodies directed against Candida in experimental models and patients with candidiasis. Because the production of antibodies against the pathogen-specific heat shock protein 90 (Hsp90) is associated with recovery from invasive candidiasis in mice and also in patients (Matthews et al., 1992, 1994), efungumab, a human recombinant antibody directed against the fungal Hsp90, has been developed. Efungumab was investigated in a double-blinded randomized multicenter study of 139 patients with invasive candidiasis (Pachl et al., 2006). Treatment with liposomal amphotericin B was compared with liposomal amphotericin B in combination with efungumab. In patients with invasive candidiasis, the combined therapy produced a significantly better clinical and culture-confirmed outcome. However, questions were raised regarding the methodology (Herbrecht et al., 2006), and additional studies are needed to establish its potential. Recently, preclinical data supporting a synergy between efungumab and caspofungin in the treatment of invasive candidiasis have been reported (Hodgetts et al., 2008). Other approaches include the use of monoclonal antibodies (mAb) and immune serum from mice that were vaccinated with Candida-mannan containing liposomes; these provided protection in mice with disseminated candidiasis (Han & Cutler, 1995; Han et al., 2000). Mice treated with recombinant anti-mannan human antibody were more resistant to disseminated candidiasis (Zhang et al., 2006). The synthetic glycopeptide vaccines can induce protective antibodies in experimental systemic candidiasis in mice (Xin et al., 2008). In addition, it is reported that β-glucan-conjugate vaccination results in anti-β-glucan antibodies, which are effective against experimental murine vaginal candidiasis (Pietrella et al., 2010).

Another approach is the use of idiotypic antibodies. Candida albicans is highly susceptible to the so-called yeast killer toxin (YKT). The monoclonal KT4 antibody neutralizes the effects of YKT. An idiotypic antibody that is directed against KT4 mAb appears to mimic the biological function of YKT (Polonelli et al., 1993), and exerts anti-candidal effects and protects against mucosal and systemic experimental candidiasis (Polonelli et al., 1993, 1994). Treatment with such antibodies that are directly effective against Candida may become a therapy for the immunocompromised host.

Cytokine therapy

Cytokines are able to enforce host defense and may therefore be useful for immunomodulation during infections. GM-CSF accelerates hemopoiesis of myeloid cells, resulting in the production of monocytes and neutrophils (Gadish et al., 1991), and in pharmacological doses, it leads to monocytosis and neutrophilia. It enhances phagocytosis and the release of ROS by PMNs (Richardson et al., 1992), and prolongs the survival of neutrophils by inhibition of programmed cell death (Brach et al., 1992). It also upregulates dectin-1 expression on macrophages (Willment et al., 2003) and promotes fungicidal activity via upregulation of chititriosidase, which cleaves chitin present in the inner cell wall of C. albicans (van Eijk et al., 2005). In neutropenic mice with disseminated candidiasis, GM-CSF was shown to reduce lung damage and mortality (Lechner et al., 1994). In two small studies in patients, Vazquez and colleagues investigated the use of GM-CSF as an adjunctive drug for clinically refractory mucosal candidiasis in patients with advanced AIDS. GM-CSF, in combination with antifungal therapy, appeared to lead to clinical and mycological improvement without adverse events (Vazquez et al., 1998, 2000). Anecdotal reports in the literature of patients with disseminated candidiasis showed a favorable response to addition of GM-CSF (Rokusz et al., 2001; Dignani et al., 2005). A case report of a patient with a 17-year history of severe CMC who was treated with GM-CSF had a favorable response (Shahar et al., 1999). Unfortunately, the patient had a severe anaphylactic reaction, which is an uncommon side effect of GM-CSF, and GM-CSF was stopped (Shahar et al., 1999). Although these studies indicate that GM-CSF could be beneficial in the treatment of Candida infections, the experience is limited and controlled trials are lacking.

G-CSF is a hematopoietic growth factor that selectively promotes the proliferation and differentiation of neutrophils. Incubation of PMNs from healthy volunteers in vitro with G-CSF showed enhanced antifungal activity in damaging Candida pseudohyphae (Roilides et al., 1995). G-CSF administered in humans also significantly enhanced PMN-mediated damage of Candida pseudohyphae (Gaviria et al., 1999). Furthermore, mice with disseminated candidiasis treated with recombinant G-CSF show significantly reduced mortality and lower fungal outgrowth (Graybill et al., 1995; Kullberg et al., 1998, 1999); this benefit was less obvious during subacute or chronic candidiasis (Kullberg et al., 1999). In chronic gastrointestinal candidiasis in mice, G-CSF, in combination with fluconazole, did not show additional benefit over fluconazole alone in reducing the fungal burden (Clemons & Stevens, 2000). The first randomized placebo-controlled trial addressing adjunctive immunotherapy in non-neutropenic patients with disseminated candidiasis compared fluconazole alone with fluconazole and G-CSF. This phase 2 study indicated that administration of G-CSF is safe and showed a trend toward faster resolution of infection (Kullberg et al., 2004).

IFNγ is produced by T cells and NK cells and augments the cytotoxic function of macrophages and the killing of intracellular pathogens (Hubel et al., 2002). IFNγ is also known to exert activity on other important non-immune cells important in host defense, such as endothelial cells, epithelial cells and fibroblasts (Gallin et al., 1995). Clinical experience with IFNγ therapy is greatest in patients with chronic granulomatous disease. In these patients, it reduces the incidence of infections, including infections with Aspergillus (Gallin et al., 1991). The effector mechanisms that are triggered by IFNγ are elusive. Several in vitro and in vivo studies support that IFNγ treatment is beneficial in the treatment of Candida infections. Various studies have shown that IFNγ increases the anti-candidal function of macrophages (Brummer et al., 1985, 1991; Brummer & Stevens, 1989; Marodi et al., 1993; Redmond et al., 1993; Baltch et al., 2005). Other studies, however, were not able to show that IFNγ enhanced the capacity of murine macrophages (Marcil et al., 2002) or murine pulmonary macrophages to kill Candida (Brummer & Stevens, 1987). Peritoneal and peripheral blood PMNs from IFNγ-treated mice showed enhanced killing of Candida (Kullberg et al., 1993). Incubation of human PMNs with IFNγ augments the capacity of PMNs to kill Candida (Djeu et al., 1986). The administration of IFNγ reduces the fungal burden in mice with disseminated candidiasis (Kullberg et al., 1993). However, IFNγ failed to improve the efficacy of fluconazole in a murine model of experimental oral mucosal candidiasis (Clemons & Stevens, 2000). Although the literature is controversial regarding the role of IFNγ in anti-Candida host defense, a small study of three patients with disseminated candidiasis reported that additional IFNγ therapy was beneficial (Dignani et al., 2005). Furthermore, administration of IFNγ in an HIV-infected patient with azole-resistant OPC resulted in a drastic improvement (Bodasing et al., 2002). These case reports suggest that IFNγ could be beneficial as adjuvant antifungal therapy, but clinical trials are urgently needed to establish whether IFNγ is valuable in the treatment of Candida infections.

Adoptive transfer of primed immune cells

Another approach for antifungal immunotherapy would be the use of antigen primed DCs that are able to skew the adaptive immune response toward anti-Candida effector functions (Bozza et al., 2004). DCs could be primed ex vivo with antigens that induce specific cytokine profiles, and thereafter infused in the patient with Candida infection (Bacci et al., 2002). It has been shown that Th1-dependent antifungal protection could be induced by DC vaccination in mice that received allogeneic bone marrow transplants (Bozza et al., 2003). Furthermore, adoptive transfer of anti-Candida T cells has been proposed as potential immunotherapy in patients with Candida infection after hematopoietic stem cell transplantation (Tramsen et al., 2007). The human T cells generated were able to damage hyphal forms of Candida and significantly enhanced hyphal damage induced by human neutrophils. As the generated T cells do not seem to be affected by cryopreservation (Tramsen et al., 2007), there is the opportunity to generate anti-Candida T cells before the patients reach an immunocompromised status, and adoptively transfer these cells during infection when patients are immunocompromised. Such new approaches to modulate the immune response offer elegant opportunities to enforce the immune system at the core of its failing anti-Candida defense mechanisms. However, the efficacy and potential adverse effects still have to be assessed in both animal models with experimental Candida infections and ultimately in the patients.

Concluding remarks

Candida infections account for a high burden of morbidity and mortality. New therapeutic approaches are urgently needed to improve the outcome of the patients, as the currently available treatment options have not reduced the mortality and morbidity associated with Candida infections over the recent years. One solution would be the use of immunotherapy, which aims at improving host defense against Candida. The increase in understanding the mechanisms that underlie the pathogenesis of Candida infection brings the development of efficient and feasible immunotherapeutic strategies closer.

Footnotes

  • Editor: Martin Kupiec

References

View Abstract